Online Principal Component Analysis in High Dimension: Which Algorithm to Choose?

Principal component analysis (PCA) is a method of choice for dimension reduction. In the current context of data explosion, online techniques that do not require storing all data in memory are indispensable to perform the PCA of streaming data and/or massive data. Despite the wide availability of recursive algorithms that can efficiently update the PCA when new data are observed, the literature offers little guidance on how to select a suitable algorithm for a given application. This paper reviews the main approaches to online PCA, namely, perturbation techniques, incremental methods and stochastic optimisation, and compares the most widely employed techniques in terms statistical accuracy, computation time and memory requirements using artificial and real data. Extensions of online PCA to missing data and to functional data are detailed. All studied algorithms are available in the  package onlinePCA on CRAN.     

Innovation sourcing excellence: Three purchasing capabilities for success

Innovation sourcing has become more critical across many industries. As global value chains have become more fragmented, change and opportunity comes from all sides. As a result, companies need to excel at capturing innovation opportunities with existing and potential supply chain members. This article describes a simple framework with three essential innovation sourcing capabilities needed to excel in purchasing: (1) Purchasing needs to explore unmet needs and anticipate future competitive advantages by working closely with other functions and clients; (2) it needs to explore external opportunities beyond first-tier suppliers; and (3) it needs to involve suppliers in innovation projects that consistently deliver results over time. Our framework has been developed based on a combined qualitative and quantitative research methodology that takes into account practices and results at the purchasing and company levels. The framework will help C-level managers and purchasing teams benchmark their progress in innovation sourcing and understand what steps need to be taken to achieve excellence.     

The Effect of Business and Financial Market Cycles on Credit Ratings: Evidence from the Last Two Decades

We analyze the effect of business and financial market cycles on credit ratings using a sample of firms from the Russell 3000 index that are rated by Standard and Poor's over the period 1986–2012. We also examine investor reaction to credit rating actions in different stages of business and financial market cycles. We document that credit rating agencies are influenced by business and financial market cycles; they assign lower credit ratings during downturns of business and financial market cycles and higher ratings during upturns. Our study is the first to find strong evidence of pro‐cyclicality in credit ratings using a long window. We also document stronger investor reaction to negative credit rating actions during downturns. Our results confirm theoretical predictions and inform regulators.     

A market game approach to differential information economies

In this paper we recast a differential information economy as a strategic game in which players propose net trades and prices. Pure strategy Nash equilibria are strong and determine both consumption plans and commodity prices that coincide with the Walrasian Expectations equilibria of the underlying economy. The authors acknowledge support by research grants BEC2003-09067-C04-01 (Ministerio de Educación y Ciencia and FEDER), PGIDT04XIC30001PN (Xunta de Galicia) and SA070A05 (Junta de Castilla y León). JP Torres-Martínez is also grateful to CNPq-Brazil and University of Vigo for financial support. We are indebted to N.C. Yannelis for helpful comments and insights.     

The Minimal Entropy Martingale Measure (MEMM) for a Markov-Modulated Exponential Lévy Model

This paper deals with the characterization problem of the minimal entropy martingale measure (MEMM) for a Markov-modulated exponential Lévy model. This model is characterized by the presence of a background process modulating the risky asset price movements between different regimes or market environments. This allows to stress the strong dependence of financial assets price with structural changes in the market conditions. Our main results are obtained from the key idea of working conditionally on the modulator-factor process. This reduces the problem to studying the simpler case of processes with independent increments. Our work generalizes some previous works in the literature dealing with either the exponential Lévy case or the exponential-additive case.     

Options as a predictor of common stock price changes

Since rather novel techniques such as neural nets allow investigation of nonlinear model specification previously untested, it may be that traditional models of price formation underperform through misspecification rather than market efficiency. This paper explores whether a multilayer backpropagation model offers exploitable profit opportunities for some limited period. Using an intradaytransaction dataset obtained from the European Options Exchange (Amsterdam), We attempted to predict the return on Philips. Two neural nets are contrasted to ordinary linear regression analysis on the basis of three benchmarks (MSE, and net realized returns). An adaptively trained 33-14-1 architecture scored best on all criteria and yielded an annualized 11% return following a simple one-period trading strategy.     

Equilibrium theory with asymmetric information and infinitely many states

Radner (Econometrica 36, 31–58 1968) proved existence of a competitive equilibrium for differential information economies with finitely many states. We extend this result to economies with infinitely many states of nature. Each agent observes a public and a private signal. The publicly observed signal may take infinitely many values but, in order to get existence, we assume that private signals only take finitely many values. Actually, there is no hope to get a general existence result since Podczeck et al. (Vienna University Working Papers 2008) already proposed non-existence results. We would like to thank a referee and the Associate Editor for valuable suggestions and remarks. The actual version of the paper has benefited from their comments. This work commenced while V.F. Martins-da-Rocha and P.K. Monteiro were visiting RGEA. Carlos Hervés Beloso acknowledges support from BEC2003-09067-C04-01 (Ministerio de Ciencia y Tecnología and FEDER) and PGIDT04XIC30001PN (Xunta de Galicia). V.F. Martins-da-Rocha acknowledges partial financial support from PRONEX. P.K. Monteiro acknowledges the financial support of CNPq/Edital Universal 02/2006.     

The Focus Efficiency of U.S. Hospitals

Hospital productivity has been a research topic for over two decades. Whereas much has been learned regarding cost, technical, scale, and allocative efficiency as well as the impact that weakly disposable inputs/outputs have on hospital behavior, we expand on this research by examining size and service offering, or focus, efficiency at the metropolitan area level for US hospitals. By using an extension of the Free Coordination Hull (FCH), we are able to determine whether hospitals in our sample could become more efficient if they provided more services (reduce inefficiency due to too narrow a focus) or fewer services (reduce inefficiency due to too broad a focus). Our results suggest that findings vary among the hospital markets. This approach could be used by policy makers and managers in order to reduce costs by sharing services, reducing services in hospitals, or expanding services in hospitals. Findings from a study such as this should aid reform programs by providing more information on the sources of hospital inefficiency.